Bashpodder/Castpodder replacement development thread
Moderators: snarkout, Patrick, dann
I just started to learn python as my first language about 3 weeks ago, so if you need someone to ask a lot of questions and generally be confused about everything going on, then I'm your man!
Does being a spectator during development help one learn anything about programming? At the very least it might be fun to look around and try to figure out what you're doing, if you happen to be using subversion or SourceForge or something like that where I can look at the code.
Does being a spectator during development help one learn anything about programming? At the very least it might be fun to look around and try to figure out what you're doing, if you happen to be using subversion or SourceForge or something like that where I can look at the code.
Vim is beautiful
I agree. What I think would be cool is an app that can be run with a gui skin or from the CLI, sort of like mplayer. That would be very cool.mowestusa wrote:I give my vote for a lightweight application. I would even go for a ncurses program (which I don't believe one exists yet) with the features discussed. I like to run old hardware, and would probably keep using bashpodder if the client was too resource hungry.
Chess Griffin
Our filtering proxy here at work didn't find it all tht funny.chuck wrote:I think you might find some of this code from the castpodder wiki useful. It was the number 6 hit from a google search on castpodder.
A lot of people would disagree with you on this. Reinventing the wheel is nearly always more work.CptnObvious999 wrote:Also it is hard to take existing code and greatly change/improve upon it but if you start from scratch you know exactly how it works and where everything is. I think it would be better to start clean and can always use bits and pieces of the code if needed.
Emacs was written as an extension to existing code, not from scratch. RMS started writing GCC by looking at other projects and began with existing code although he eventually rewrote from scratch. ESR also recommends starting a project based on existing code. Fetchmail was written that way.
Sure, you have more control when you write everything from scratch, but it's certainly not easier.
- CptnObvious999
- Posts: 798
- Joined: Fri Jun 03, 2005 7:54 pm
- Location: Maryland
- Contact:
Ok, that's true but over half of all the code is specifically for the GUI. Maybe it would be easier to just take the CastPodder code but I think it would be better to just take some of the code they have and fit it into the project. But this is Pat's idea so he can do whatever he wants.Gomer_X wrote:A lot of people would disagree with you on this. Reinventing the wheel is nearly always more work.CptnObvious999 wrote:Also it is hard to take existing code and greatly change/improve upon it but if you start from scratch you know exactly how it works and where everything is. I think it would be better to start clean and can always use bits and pieces of the code if needed.
Emacs was written as an extension to existing code, not from scratch. RMS started writing GCC by looking at other projects and began with existing code although he eventually rewrote from scratch. ESR also recommends starting a project based on existing code. Fetchmail was written that way.
Sure, you have more control when you write everything from scratch, but it's certainly not easier.
Is that ever a great point. I love it when applications offer cli and gui interfaces. I don't know how much effort and time this adds to the programming process, but sometimes it drives me crazy when I can't throw something in a cron job or a script because the program only uses a gui.Chess wrote:I agree. What I think would be cool is an app that can be run with a gui skin or from the CLI, sort of like mplayer. That would be very cool.mowestusa wrote:I give my vote for a lightweight application. I would even go for a ncurses program (which I don't believe one exists yet) with the features discussed. I like to run old hardware, and would probably keep using bashpodder if the client was too resource hungry.
Vim is beautiful
Actually, it's my idea viewtopic.php?p=10967&sid=7fe38dd21f1be ... be26#10967CptnObvious999 wrote: Ok, that's true but over half of all the code is specifically for the GUI. Maybe it would be easier to just take the CastPodder code but I think it would be better to just take some of the code they have and fit it into the project. But this is Pat's idea so he can do whatever he wants.
And someone mentioned starting a subversion repository for it. I'll take care of that too. Let's put it at:
svn co svn://linc.homeunix.org/feedcatcher
Of course, you'll need to get a hold of me to be able to contribute to the project through subversion, but you should be able to download what's there (not that there is anything yet).
Now's your chance guys, Jump right in and get started! Just make sure to email me so I can get a login name and password so you can actually commit changes to the program through subversion.
-Linc Fessenden
In the Beginning there was nothing, which exploded - Yeah right...
In the Beginning there was nothing, which exploded - Yeah right...
- Spankminister
- Posts: 22
- Joined: Sun Nov 12, 2006 11:56 am
- Contact:
Seems to me that we'll need:
1) A roadmap with concrete features and milestones we can meet... We can't make progress if everyone has a different idea of what the software should do.
2) Some sort of web presence as a reference and posting place for progress.
Anything else? Also, if someone has a link to the most recent CastPodder source tarball, that would be a great reference for everyone. I have no problem with starting with CastPodder and rewriting whatever sections we have to to make it modular and extendible, but what do you guys think? I generally prefer using the CastPodder/Juice style window, but ncurses is an interesting idea as well. I can't think of any at the moment, but there must be some program that does both well?
1) A roadmap with concrete features and milestones we can meet... We can't make progress if everyone has a different idea of what the software should do.
2) Some sort of web presence as a reference and posting place for progress.
Anything else? Also, if someone has a link to the most recent CastPodder source tarball, that would be a great reference for everyone. I have no problem with starting with CastPodder and rewriting whatever sections we have to to make it modular and extendible, but what do you guys think? I generally prefer using the CastPodder/Juice style window, but ncurses is an interesting idea as well. I can't think of any at the moment, but there must be some program that does both well?
abstk
That looks slick. I do wonder if it supports a non-ncurses interface. I was thinking CLI for cron jobs and scripts, though I really don't know much about ncurses aside from using it occasionally.
That link shows the 0.1 release from May 2004. Looking at some of the Gobolinux links, it appears that they're on version 0.3, but I can't find a package for it anywhere.
That link shows the 0.1 release from May 2004. Looking at some of the Gobolinux links, it appears that they're on version 0.3, but I can't find a package for it anywhere.
Vim is beautiful
- Spankminister
- Posts: 22
- Joined: Sun Nov 12, 2006 11:56 am
- Contact:
I didn't see it hosted anywhere, so I found the Mandriva CastPodder rpm and extracted the tarball from it. You can find it here. Let me know if it doesn't work.
I'm of the opinion that job one should be getting something working with this abstk, and then fitting chunks of CastPodder code into that GUI and building from there.
I'm of the opinion that job one should be getting something working with this abstk, and then fitting chunks of CastPodder code into that GUI and building from there.
- TankCatNinjaFish
- Posts: 110
- Joined: Wed Aug 09, 2006 4:29 am
one small advice: instead of making a list of all features which you want in a the program, get a bare bone skeleton running. With python a prototype should be done in a weekend or less. In fact, the initial decisions should be made unilaterally by the founder (who is, I guess, patrick?) to avoid flame wars over technology. Arguments over basic design, class hierarchy, GUI toolkit, source code control, license can all stop a project dead in its tracks.
As for the re-inventing-the-wheel theory, I say its alright to write something from scratch when its as simple as a podcatcher.
As for the re-inventing-the-wheel theory, I say its alright to write something from scratch when its as simple as a podcatcher.